Modeling the price of CEZ shares

The bachelor's thesis focuses on the analysis and prediction of CEZ share price development between 1999 and 2024. We evaluate the basic statistical properties of the time series such as variability, homoscedasticity, normality, seasonality, stationarity and autocorrelation of residuals. We remove structures that violate the assumptions of classical models, for which we use, among other things, descriptive statistics. When making predictions, we apply trend estimation using moving averages, polynomial functions, Fourier series, Brown's double and triple exponential smoothing, and more complex methods by incorporating the ARIMA process. We express the risk according to the so-called "Value at Risk" and compare the agreement of estimates of all approaches with the share price from January to May 2025.